NuVinci no thanks

NuVinci just tried to sell me a "Developers Kit" for about $600, which they are promoting for light EV use, so I took the opportunity to ask some questions. I asked about efficiency and the answer I got was "Efficiency is in the low 90’s as a peak at 1:1". That tells me it loses 9% in direct drive and then efficiency falls off at the lower speed ratios. Add in that full up the unit and its devices weigh 11.5lbs. Now I understand why no ones raving about them. Maybe once battery energy density goes up 10 fold and prices come way down they'll find a market. Until then it's just a cool idea that almost works well enough.

John

From what I read they don't like to talk much about specifics on efficiency. I believe they have a forum on their web site if I remember correctly, and don't give numbers when asked directly, what they told you is in line with what I have heard others talking about 10-15% loss.

I'll be putting one to the test this summer using one of Recumpence's 2 stage drives with an Astro 3210 and an HV110 controller.. :lol:

What it looses in efficiency i'm hoping to gain in simplicity and ease of use.. time will tell.

I'm keen to see how well it does as well. I think the loss in efficiency might be more than made up for by the huge ratio change you get, which means gobs of torque, at the low end, and high speeds at the other extreme. I really think you'd notice the efficiency loss more on a regular pedal-only bike, or with a under-powered motor.

I think you also might want to reconsider and just get a single stage verison of Matt's drive, as you will have a lot more ratio change than most of us will with 3-speed hubs or standard cassettes.

-- Gary

What we need is a 2 speed shaft drive gearbox. Something light, simple, and durable. What's surprising is that it doesn't already exist, though the 3 speed self propelled lawnmower transmission I picked up for $20 doesn't miss the mark by too much.

John

John in CR said:

What we need is a 2 speed shaft drive gearbox. Something light, simple, and durable. What's surprising is that it doesn't already exist, though the 3 speed self propelled lawnmower transmission I picked up for $20 doesn't miss the mark by too much.

John

Click to expand...

I wholeheartedly agree. I don't think we need more than two gears on an ebike in order to get the best of both worlds, lots of low-end torque and still have a decent top speed. Otherwise you need an overvolted 25lb hub motor to get both, but it ends up weighing as much as a motorcycle.

There are plenty of small and powerful motors, but with these, unless you go through the gears or a multi-speed hub, you can't get power and speed. The retrodirect idea is interesting, but we'd need complicated phse wire switching, or a controller that can go both directions with out having to go through a setup menu. Most RC car controllers are bi-directional, but I think most of these are even more voltage limited than the "HV"-type airplane and helicopter ESCs.

A simple two-speed transmission would make things a lot less complex.

-- Gary

GGoodrum said:

The retrodirect idea is interesting, but we'd need complicated phse wire switching, or a controller that can go both directions with out having to go through a setup menu. Most RC car controllers are bi-directional, but I think most of these are even more voltage limited than the "HV"-type airplane and helicopter ESCs.

Click to expand...

I was thinking of these:
http://www.castlecreations.com/products/hydra_hv_line.html

Last Friday I got to test ride Justin's Big Dummy with the NuVinci hub. It was a revelation! What a nice piece of equipment. It's smooth and silent. It can't slip out of gear the way an internally geared hub does from time to time so it inspires confidence when mashing up a hill. That you can still shift while you're mashing is what I found remarkable.
He plans to eventually couple it with a Schlumpf-two-speed-crankset. That will make for a simple and elegant drive train with all the range one could use.

Quibbling about a couple extra kilos on an ebike is a non-starter, IMO.

Miles said:

GGoodrum said:

The retrodirect idea is interesting, but we'd need complicated phse wire switching, or a controller that can go both directions with out having to go through a setup menu. Most RC car controllers are bi-directional, but I think most of these are even more voltage limited than the "HV"-type airplane and helicopter ESCs.

Click to expand...

I was thinking of these:
http://www.castlecreations.com/products/hydra_hv_line.html

Click to expand...

I ran one for a while and took it off. The timing advance is too high for low KV outrunners. So, it loses sync easily under load.

For your retro direct, I was thinking about a relay setup that reverses two motor wires to reverse motor direction. A simple throttle off sensor could be used to assure the relay system was not engaged with throttle on. So, you throttle off, flip the reverse switch, then hit the throttle and you are reversed and into the next ratio.

Matt

Although it is nice that they have "HV" versions of the Hydra series, like Matt says, the S/W might be an issue. I've tried to get them to make slight software changes before, like adding the softstart feature so we don't keep stripping main gears on big RC helicopters, but it took them 2-1/2 years to do that so I don't think we can count on getting them to make a change to the timing that isn't already part of the setup menu.

I do think the R-D idea is worth looking at though, especially if we could do it with belts in a "growth" version of Matt's drive.

--Gary

Zoot,

I appreciate that it would be nice and smooth and could make a Cadillac of a bike, but batteries are the single biggest cost of an ebike. Anything that coughs up 10% of my batteries is a non-starter. Hub motors prove that single speeds just aren't that effective except with flat terrain, but some do get pretty close to an acceptable range except for their low speed efficiency hit. A 2 speed gearbox is simple enough, and would be sufficient for most cases due to the wide efficiency band of electric motors, so it's really not asking too much at all. The advanced design of the NuVinci is a great piece of engineering, but it's overkill for what we need, and that kind of efficiency hit isn't acceptable to me, just like I wouldn't switch to a 75% efficient motor even if it could climb hills. This is a case where the simplest form will prove best, I just find it hard to fathom that with the Chinese putting 20 million new 2 wheel EVs on the road each year that the answer doesn't already exist in a cheap and simple form.

Also, since NuVinci isn't talking except that the efficiency hit is close to 10% at the unit's peak efficiency, I'd lay pretty good odds that the efficiency hit in the lowest ratio is at least another 10% hit and possibly another 15-20% or more. That would explain why the gentleman with an Etek took his NuVinci off, and now has better acceleration with only a slight loss of top end speed. I'm sorry but an Etek with gearing should have so much jump that the problem is keeping the front wheel on the ground, not lackluster takeoffs.

John

The bike has an E-Zee front hub motor so wheelies are non-starter.
It makes for a good combination with the NuVinci.

Zoot Katz said:

The bike has an E-Zee front hub motor so wheelies are non-starter.
It makes for a good combination with the NuVinci.

Click to expand...

Wait, the NuVinci is for pedaling only? I guess that's why the bike is named Big Dummy. If I'm pedaling, I'm definitely not coughing up 10% for nothing, plus another 10-20% going up hills. I guess the motor on the front is mandatory, to help you pedal the darn thing.

John

John in CR said:

Zoot Katz said:

The bike has an E-Zee front hub motor so wheelies are non-starter.
It makes for a good combination with the NuVinci.

Click to expand...

Wait, the NuVinci is for pedaling only? I guess that's why the bike is named Big Dummy. If I'm pedaling, I'm definitely not coughing up 10% for nothing, plus another 10-20% going up hills. I guess the motor on the front is mandatory, to help you pedal the darn thing.

John

Click to expand...

I don't see why the brains at NuVinci couldn't move the thing with a shaft and right angle gear drive.
What happened to your idea of using a grinder gear box?
You can lose as much efficiency to drag just wearing baggy shorts and a flapping wind-breaker.

The NuVinci is a simple and elegant drive for pedal powered bicycles. I hope the company succeeds. They need a carbon fibre shell for the weight weenies.

.

It really can be a hub motor. Cool drive train. Just needs a stator and magnets and a Chinese "build".

Zoot Katz said:

What happened to your idea of using a grinder gear box?

Click to expand...

It's in the line of projects. I do have some shaft adapters for 3 different motors and a mounting plate already made. I've got a couple of the gearboxes and just need to go on a search for suitable gears to use. I call them gearboxes because they enclose a pair of helical gears, but all they are is a right angle gearset with a 3 or 4 to 1 reduction. It doesn't solve the issue of needing multiple speeds.

I also talked to a company that makes bicycle shaft drives, but the pricing was too rich for my blood, especially considering there was a question of durability with the kind of input power of a 1kw or more electric motor.

John

GGoodrum said:

I do think the R-D idea is worth looking at though, especially if we could do it with belts in a "growth" version of Matt's drive.

Click to expand...

I'm still working on it, Gary :wink:

I rode a thing built by Thunderstruck Motors that had a NuVinci hooked up to a 5kw or so custom motor. It was kinda cool, but according to tests done with the NuVinci and without it (single fixed gear), the efficiency gained by having variable ratios was pretty much cancelled out by the losses in the transmission for normal riding.

I used a NuVinci in an ICE application and found that the gearing would not actuate well under torque. I think some others here reported the same thing. I was using a Honda GXH50 motor through gears giving about 60-70 Nm torque at the hub.

The housing is aluminum, so most of the weight is from the steel spheres inside. They would have to find another material for the spheres to get serious weight reductions.

I'm surprised they don't use ceramic balls, or is that not where all the weight is? Anyway, it's a good concept but seems to fall on it's face in the real world, or at least in this application.

When is somebody going to cram one of these things (or a Shimano 3 speed) inside a hub motor? It seems rather obvious to me, although I don't know how it would work, exactly. That would make me switch back to hub.

Anything with >10% drivetrain losses is going to be building up a whole of lot heat. 200w of heat from just 2000w of motor when at it's most efficient. Seems strange they can't make the 1:1 into something that locks up solid to give 100% efficiency.

That said, compared to hub motor guys dealing with only 20-30% efficiency when at low rotor speeds, having a 10-20% loss that bumps your motor efficiency to a rotor speed point that is in a more efficienct range could easily more than compensate for the mechanical losses. That would improve total efficiency, along with a drastic improvement in acceleration performance. However, this isn't for hub motors, so it's a mute point.

It might not be the ultimate best ever solution, but it's a step in the right direction. Nothing wrong with stepping in the right direction, and when the internally geared hub systems mature a bit, it could be pretty slick.

liveforphysics said:

. . .

It might not be the ultimate best ever solution, but it's a step in the right direction. Nothing wrong with stepping in the right direction, and when the internally geared hub systems mature a bit, it could be pretty slick.

Click to expand...

Mature? They've already had 114 years of service and development.
Its seems strange that a patent for first internally geared hub was granted to a machinist who also hailed from Indiana. . . In 1895. His design was not commercially successful. It was broadly copied three years later and remained in production for a decade proving that compact epicyclic hub gears were practical for a bicycle.

The NuVinci would not be my hub of choice for racing because of the weight/efficiency hit.
But for bomb squad duty on a city bike, it's the most elegant answer to date.

Zoot Katz said:

...But for bomb squad duty on a city bike, it's the most elegant answer to date.

Click to expand...

Agreed, but I believe that's only because no one ever focused on a simple 2 speed geared hub for an ebike. The mechanical engineers aren't interested because it's too simple, and their bosses aren't interested because they don't see a big enough market. What I can't get past is why the Chinese haven't come up with it yet. 20 million+ new 2 wheel EVs a year would seem to stimulate creativity. Imagine what would be available to us if the US or Europe were putting 20 million new 2 wheel EVs on the road each year.

John

I think making a hub with a single set of planetary gears inside running from the axel, and then something that just locks the axel to the hub solidly would be both efficient, light weight, and fairly simply to make. Something like 4:1 planetary gear reduction for acceleration, and then 1:1 for cruising.

I doubt it would be too tough to do, but I personally wouldn't want to try making it

I saw a video on youtube of a hubmotor setup that ran gears for starting, and then locked 1:1. I think it had an electric switch or something you stuck on the handle bars. But I think it was something in the very weenie power range.

A powerful hub motor with a 2-speed that was under 15lbs would make me switch to a hub.